Research Article
[Retracted] Active Learning Query Strategies for Linear Regression Based on Efficient Global Optimization
Table 3
Percent improvements of the AUCs of the mean RMSEs, CCs, and Oppo Costs compared with each original ALR approach.
| | | EGO-EMCM | EGO-QBC | EGO-GSx | RD-EGO |
| Ridge | RMSE | −0.96 | −0.29 | −1.07 | 1.31 | CC | −0.16 | −0.27 | −0.45 | −0.13 | Oppo Cost | 15.37 | 13.42 | 14.57 | 54.37 | Lasso | RMSE | −0.99 | −0.81 | −0.82 | 3.48 | CC | 1.20 | −0.36 | −0.48 | −0.90 | Oppo Cost | 22.17 | 10.92 | 16.35 | 55.45 | Enet | RMSE | −0.70 | −0.68 | −0.75 | −0.88 | CC | 0.37 | −0.55 | −0.62 | −1.99 | Oppo Cost | 18.60 | 11.82 | 15.82 | 54.40 |
|
|